Sign in

Eric

@edavis.dev
2.6K followers 969 following 2.4K posts
PostsRepliesMedia
Eric @edavis.dev · 13h
thank you! yeah that makes sense
010
Eric @edavis.dev · 13h
gotcha, thank you!
010
Eric @edavis.dev · 15h
can a LLM knower explain to me what changes between "model names" (eg, Sonnet -> Opus -> Fable) and what changes between "model versions" (eg, 5 -> 5.5)? Like do the names represent fundamentally different parameter counts and the versions are more incremental improvements?
251
Eric @edavis.dev · 27/09/2026
ah nice I love Dune
121
Eric @edavis.dev · 26/09/2026
v-it.org / @v-it.org (haven't used it myself so can't vouch, but looks cool)
v-it.org
vit — open source is social
developers and AI agents share software capabilities through a social network — discover what others have built, adapt it to your codebase, build reputation by vouching for what works.
170
Eric @edavis.dev · 23/09/2026
how about onlyelders.com
060
Eric @edavis.dev · 16/09/2026
hell yeah
010
Eric @edavis.dev · 13/09/2026
a HELP PEER knuckle tattoo would go hard
5635
Eric @edavis.dev · 13/09/2026
I got a good feeling, I think it'll happen
041
Eric @edavis.dev · 07/09/2026
this took all my remaining IQ points
OP: "That list included like 3 Black ppl."

Reply: "is that related to the use of claude code?"

Different person: "Signs point to yes. Sadly."
1180
Eric @edavis.dev · 06/09/2026
🚨🚨🚨HAD PIZZA FOR DINNER🚨🚨🚨 PUT SOME CRUSHED RED PEPPERS ON IT hope everyone is having a nice saturday ❤️❤️❤️
0230
Eric @edavis.dev · 05/09/2026
the kicker is I even paused before typing it, but convinced myself them being at the end and part of a string would make it okay but yeah, grade A name, perfect for it
030
Eric @edavis.dev · 05/09/2026
this is the second time it's happened, first was @2048.blue so @pds.dad is now 2/2
130
Eric @edavis.dev · 05/09/2026
ha, so this was my mistake. I initially was emitting "com-bsky38" but labels can't have numbers so they're recorded here but Bluesky never ingested/displayed them "com-bskythirtyeight" is the workaround dashes are a munin thing, dots break things
120
Eric @edavis.dev · 05/09/2026
the new thread count badges are cool but seems like they should be using non-breaking spaces? otherwise you can get " X/Y" on the final line
000
Eric @edavis.dev · 04/09/2026
big time
screenshot of @recordcollector.edavis.dev emitting a bunch of com-bsky38 labels
250
Eric @edavis.dev · 29/08/2026
> modern Linux distributions have dropped the legacy, deprecated PCRE1 (libpcre3) library in favor of the newer PCRE2 standard 1 < 2 > 3
020
Eric @edavis.dev · 28/08/2026
you can never be too careful
000
Eric @edavis.dev · 28/08/2026
made sure to wear my eclipse glasses tonight
100
Eric @edavis.dev · 28/08/2026
no it was this app, but you may have answered my question because I saw it on pdsls too I assumed it was scoped to a certain space, but I can see how this would generate that warning with my limited oauth scope understanding
wot.sky.boo scope listing screen
010
Eric @edavis.dev · 28/08/2026
this language confuses me: > This application is requesting access to every space on the network. This is a very broad permission. Only grant it to applications you deeply trust. I would have expected something more like "... to every record in this space" where is my intuition letting me down?
100
Eric @edavis.dev · 27/08/2026
big day for toggles on the Privacy and Security settings screen
010
Eric @edavis.dev · 25/08/2026
she's on a generational heater right now folks, let your kids stay up late to watch this one
1221
Eric @edavis.dev · 25/08/2026
jetstream1.us-west.bsky.network for that one
040
Eric @edavis.dev · 23/08/2026
figproto!
010
Eric @edavis.dev · 21/08/2026
the atproto urge to rebuild existing sites will go into hyperdrive with spaces
3414
Eric @edavis.dev · 21/08/2026
aturi.to has spaces support
150
Eric @edavis.dev · 21/08/2026
vͩo̔̈tͦͪ̓ḯ͖͆͜n̯̒͝ͅg̱̪͋̌ ͙͉̗t̵̥͗ḯ̝m͙e
020
Eric @edavis.dev · 20/08/2026
figuring out where "spreadsheet people" originated by tracking every usage I encounter in a — you guessed it — word doc
090
Eric @edavis.dev · 20/08/2026
they're born furled
020
Eric @edavis.dev · 20/08/2026
well well well if it isn't the consequences of my own actions
screenshot of Slack settings with "Show text previews of linked websites" unchecked
040
Eric @edavis.dev · 20/08/2026
bsky.app post links should unfurl in Slack, right? there's not a known limitation there or anything?
220
Eric @edavis.dev · 17/08/2026
a recent reply to a 4 month old post about AI and outages surfaced it (atproto feed!) I was going to quote but thought better of it
150
Eric @edavis.dev · 17/08/2026
do people think Bluesky adds AI code, it breaks the app, they remove the code, that fixes things and then the cycle repeats next time AI is used?
6692
Eric @edavis.dev · 16/08/2026
mvkel 5 months ago | parent | context | favorite | on: Digg is gone again
The patterns were there if you knew to look for them.
The original Digg excepted, Kevin Rose's attention span is extremely limited. He will give something ~3-4 months of attention before (apparently) getting bored and wanting to move on to something else.
Up until that point, he will be an unrelenting hype man of whatever his attention is lasered on at that moment.
Then the hype posts start to drift. They show up once every few days, then once a week, then stop entirely. Any criticism or skepticism is considered a buzz kill in the cloud of good vibes only.
A few months later, a dramatic explainer post arrives (underestimating the cold start problem?
Really??), outlining why the idea didn't work and why the next one will be better, for sure, for real.
This (AI generated) note from the current CEO paints an optimistic picture, but the most likely outcome will be that Digg simply doesn't launch.
It's sustained on the nostalgic vapors of the old guard, not renewed by a replenished sense of purpose, or connection.
I'd say I'd love to be proven wrong, but I personally question the utility of a Web 2.0 social network phoenixing itself. We have endured a decade+ of originality being buffed out of web products, most now resembling variations of Bootstrap and shaden in service of dev convenience and getting rich quicker.
Surely in the age of vibe coding, we can afford to take creative risks again, and think of something new.
180
Eric @edavis.dev · 16/08/2026
sorry for blowing up your spot! I only shared it in the spirit of "looks like bluesky is cooking up something here" looking forward to seeing where it goes!
010
Eric @edavis.dev · 16/08/2026
trying to find something, coming up empty so far: a dev (I think one employed by bsky, but not positive) talking about how "Lexicon RPC" is the essence of atproto in a lot of ways and how the new @atproto/lex package makes this construct very easy to build anybody else remember seeing this?
130
Eric @edavis.dev · 16/08/2026
I hadn't seen it shared, just noticed it on github.com/bluesky-social
2120
Eric @edavis.dev · 16/08/2026
👄
210
Eric @edavis.dev · 16/08/2026
account.info 👀
account.info
account.info — AT Protocol profile lookup
Fetch AT Protocol profiles and avatars.
5476
Eric @edavis.dev · 15/08/2026
think of it as "xrpc.v1.json" in full rather than focusing on the version bit it's the defined format for Jetstream's subscribeEvents endpoint legacy Jetstream's /subscribe didn't use this subprotocol concept the Relay's subscribeRepos uses "xrpc.v0.cbor"
020
Eric @edavis.dev · 11/08/2026
w social is the first atproto-native mastodon instance on the network
3504
Reposted by Eric
ANSI Files from Discmaster @ansibot.bsky.social · 08/08/2026
KI-THING.ANS, 1996-09-30 from dark domain - the artpacks.acid.org collection (CD-ROM)
thing (c) marvel enterta: inment group - - NO - - : : - L - - - - - - - - - - - -
071
Eric @edavis.dev · 01/08/2026
running the reference PDS with @atproto/pds pointing to the permissioned-data branch... spaces spotted 👀
screenshot of `.tables` output from a store.sqlite including tables prefixed "space"
1131
Reposted by Eric
Aaron Ross Powell @aaronrosspowell.com · 29/07/2026
This post got me writing. www.aaronrosspowell.com/3mrsuois6422y
aaronrosspowell.com
Your Internet Doesn't Need to Be Big
But it definitely can be better.
1254
Eric @edavis.dev · 29/07/2026
@freek.bsky.social not super active though
010
Eric @edavis.dev · 21/07/2026
[reading the jacobian conjecture counter-example but not realizing I'm holding it upside down] this is amazing
0182
Reposted by Eric
Kuba Suder 🇵🇱🇺🇦 @mackuba.eu · 20/07/2026
I'm starting a new blog post series, with some more practical tutorials about ATProto things, using Ruby and my Ruby gems :] Episode 1 - resolving handles & DIDs and plc.directory:
mackuba.eu
ATProto in Practice #1: Identity
I’m starting a new blog post series that I’ve named “ATProto in Practice”. I want to go through some practical tasks that you’re likely to need when working with the protocol, using Ruby and my Ruby gems for the examples. These will (hopefully) be a bit shorter than my standard book-length blog posts here 🫠 If you’re new to the AT Protocol and you haven’t read my long Introduction to AT Protocol post that goes through the whole architecture and defines the various pieces of it, I recommend you read that one first, because I won’t be explaining everything from scratch again here. Let’s start with something pretty fundamental: user identity. DIDs & handles Every account has an immutable identifier called a DID (Decentralized Identifier), e.g. did:plc:z72i7hdynmk6r22z27h6tvur. This ID serves a similar role as an UUID might in an SQL database – it’s used everywhere behind the scenes in the protocol for any references between accounts and records, in URIs and so on. But generally every account also has a human-readable handle assigned to it, which can be changed at any time, and this handle is normally what you see in the UI, often in user-facing URLs, what you use to log in with, etc. The handle is a domain name, e.g. @python.org, and any existing domain name you own and use for your website can be used as your handle. As you start building on ATProto, one of the first things you might need to do is convert between these two identifiers. For example, a user logs in to your app using a handle and you need to first “resolve” it to a DID, or a URI or other reference in a record points to a DID, but you want to show the corresponding user’s handle in the UI, and the handle is not included in the data you got. A useful UI tool for quickly looking this up is internect.info. If you look up e.g. “firefox.com” there, you will get a page showing details of Firefox’s ATProto account: the DID (did:plc:m424cqoxwhxgjutbta7jmrur), when it was created, assigned PDS hostname, and so on. Now, how to do the same thing in code? Every DID has a so called DID Document, a JSON file storing some basic info about that identity, and among other things, that document includes the alsoKnownAs field, which lists the assigned handles, in the form of at:// URIs with only the first segment. Almost every account will have exactly one handle assigned, but it might happen that one will have zero or more than one. It might also happen that the array will contain some invalid strings which aren’t proper handles – e.g. don’t include the at:// prefix, or only contain one word and no periods. You need to filter only the valid ones. The assignment is bi-directional – a DID has one (usually) handle assigned in its document, and a handle resolves to a DID. You should ideally check the assignment in both directions, because it could happen that the other sign of the assignment is no longer valid, or worse – has never been real. That is, when you resolve a handle to a DID, check if th…
37610
Eric @edavis.dev · 20/07/2026
bsky.app/profile/edav...
070
Eric @edavis.dev · 20/07/2026
Hi I'm Bluesky Social, welcome to sports app
0210